Determining Cadmium Critical Concentrations in Natural Soils By

Determining cadmium critical concentrations in natural soils by assessing Collembola mortality, reproduction and growth Thomas Bur, Anne Probst, Audrey Bianco, Laure Gandois, Yves Crouau

To cite this version:

Thomas Bur, Anne Probst, Audrey Bianco, Laure Gandois, Yves Crouau. Determining cadmium critical concentrations in natural soils by assessing Collembola mortality, reproduction and growth. Ecotoxicology and Environmental Safety, Elsevier, 2010, Vol. 7, pp. 415-422. 10.1016/j.ecoenv.2009.10.010. hal-00905962

HAL Id: hal-00905962 https://hal.archives-ouvertes.fr/hal-00905962 Submitted on 19 Nov 2013

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés.

=.-& M*A"#/- !)'1)'$- M*A"#/- ='/-*(- R=M!M=S =M!M= #$ %& ).-& %AA-$$ *-.)$#()*N ("%( A)11-A($ ("- Y)*Z )5 !)'1)'$- *-$-%*A"-*$ %&, K%Z-$ #( 5*--1N %/%#1%01- )/-* ("- Y-0 Y"-*- .)$$#01-4

!"#$ #$ %& %'(")*+,-.)$#(-, /-*$#)& .'01#$"-, #&2 "((.233)%(%)4'&#/+()'1)'$-45*3 6.*#&($ 782 9:;<

!" #$%& '" '($) *+'$,#-2 8=72>?4>?>:3@4-A)-&/4B??C4>?4?>? DEF2 "((.233,G4,)#4)*H3>?4>?>:3@4-A)-&/4B??C4>?4?>?

!" ,$'- '($) .-+)$"% 2 I'*J !")K%$ %&, L*)0$(J M&&- %&, I#%&A)J M',*-N %&, O%&,)#$J F%'*- %&, P*)'%'J Q/-$ RB?10S ! ! "#$%$%&'()*#$+#' ("$!$(),'(-%( %!")!$-%.'$%'%)!+"),'.- $,.'/0').. ..$%&'1-,, #/-,)'#-"!),$!02' " 3"-*+(!$-%')%*'&"-4!56 6A)()G#A)1)HN %&, 6 &/#*)&K-&(%1 T%5-(NJ U)14 <9 R& 9S4 ..4 ;>W+;BB4 7TTX ?>;<+:W>9

M&N A)**-$.)&,-&A- A)&A-*&#&H ("#$ $-*/#A- $")'1, 0- $-&( () ("- *-.)$#()*N %,K#&#$(*%()*2 $(%55+)%(%)[#&.+()'1)'$-45* Determining cadmium critical concentrations in natural soils by assessing Collembola mortality, reproduction and growth$

T. Bur a,c, A. Probst a,c, A. Bianco a,c, L. Gandois a,c, Y. Crouau b,c,n a Universite´ de Toulouse, UPS, INP, EcoLab (Laboratoire d’e´cologie fonctionnelle), ENSAT, Avenue de l’Agrobiopole,ˆ F-31326 Castanet-Tolosan, France b Universite´ de Toulouse, UPS, INP, EcoLab (Laboratoire d’e´cologie fonctionnelle), 118 route de Narbonne, F-31062 Toulouse, France c CNRS, EcoLab, F-31326 Castanet-Tolosan, France

a b s t r a c t

The toxicity of cadmium for the Collembola Folsomia candida was studied by determining the effects of increasing Cd concentrations on growth, survival and reproduction in three cultivated and forested soils

with different pH (4.5–8.2) and organic matter content (1.6–16.5%). The Cd concentration in soil CaCl2 exchangeable fraction, in soil solution and in Collembola body was determined. At similar total soil concentrations, the Cd concentration in soil solutions strongly decreased with increasing pH. Reproduction was the most sensitive parameter. Low organic matter content was a limiting factor Keywords: for reproduction. Effect of Cd on reproduction was better described by soil or body concentrations than Ecotoxicity by soil solution concentration. Values of EC50-Repro expressed on the basis of nominal soil concentration Soil À were 182, 111 and 107 mg g 1, respectively, for a carbonated cultivated soil (AU), an acid forested soil Collembola with high organic matter (EPC) and a circumneutral cultivated soil with low organic content (SV). Cadmium Reproduction Sensitivity to Cd was enhanced for low OM content and acidic pH. The effect of Cd on reproduction is Mortality not directly related to Cd concentration in soil solution for carbonated soil: a very low value is found for Growth EC50-Repro (0.17) based on soil solution for the soil with the highest pH (AU; pH=8.2). Chronic toxicity À Bioaccumulation cannot be predicted on the basis of soluble fractions. Critical concentrations were 8 Â 10 5, 1.1, Critical load 0.3 mg mLÀ1, respectively, for AU, EPC and SV soils. pH

1. Introduction Bakker, 1996; Probst et al., 2003a, b; Slootweg et al., 2005). CL are based on the estimate of the highest concentration in soil solution Critical load (CL) was defined as ‘‘the quantitative estimate of (critical concentration, CC, De Vries et al., 2007) supposed to have an exposure to one or more pollutants below which significant no effect on a soil function, community or population. Indeed this harmful effects on specified sensitive elements of the environ- CC remains difficult to determine. Actually, models based on free ment do not occur according to the present knowledge’’ (Nilsson ions concentrations are often used for that purpose (Lofts et al., and Grennfelt, 1988). CL has first been developed to evaluate and 2004). But calibrations are still needed regarding toxicological mitigate the effect of acid rain on forest ecosystems in Europe (De impact on various soil organisms under different soil properties Vries, 1988; De Vries et al., 1994; Warfvinge and Sverdrup, 1992; (De Vries et al., 2007). Bioassays using soil organisms might be Hornung et al., 1995) and particularly in France (Dambrine et al., useful tools to determine CC if they refer to ecologically relevant 1993; Massabuau et al., 1995; Party et al., 1995, 2001; Moncoulon parameters like mortality, growth or reproduction. Molecular et al., 2004, 2007). Since 1991, CL maps are used for international parameters (for example enzymatic activities) do not allow to negotiations on air pollution abatement strategies (Hettelingh determine the ecological impact of a specific chemical. Five et al., 1991). It is now well admitted as a powerful tool to estimate standardized tests using soil animal models are used in Europe: ecosystem sensitivity to pollutant inputs. Eisenia fetida (Annelide) mortality and reproduction tests (Inter- As a first attempt, CL for heavy metals (HM) in soils (De Vries national Standard Organisation, ISO 11268-1, 1994, ISO 11268-2, and Bakker, 1996) were estimated in a similar way (De Vries and 1997), Folsomia candida (Collembola) reproduction test (ISO 11267, 1998), Enchytraeids reproduction test (ISO 16387, 2001), and the assay for field testing with earthworm (ISO 11268-3, $ The experiments were conducted in accordance with national and institu- 1999). These assays usually concern normalised substrate rather tional guidelines for the protection of human subjects and animal welfare. n than field soils. Corresponding author at: Universite´ de Toulouse, UPS, INP, EcoLab (Laboratoire d’ećologie fonctionnelle), 118 route de Narbonne, F-31062 Toulouse, France. Collembola are relevant target organisms to determine CC of E-mail address: [email protected] (Y. Crouau). HM in soils since (i) they are detritivorous, contribute to the ARTICLE IN PRESS

nutrient cycle in soils and are widespread in most natural soils. (particularly the carbonated soil, AU) compared to the acidic forested soil (EPC). Moreover, they have been extensively studied, particularly Soil moisture were set up to 25%, 30% and 20%, respectively, for AU, EPC and SV (50% water holding capacity, WHC), to enhance development of Collembola by F. candida, which is commonly used as a biological model in creating an adequate crumbly structure. Cd concentrations were in the range of laboratory test for pollutant toxicity assessment. The F. candida low contaminated soils (Baize, 1997). reproduction test is increasingly used because it is sensitive and The aim of the study was to compare the responses to Cd toxicity of various because the breeding of F. candida is easy (Riepert, 1995). It was soils with different physico-chemical characteristics. Consequently, to match field conditions as closely as possible, the ISO 11267 guidelines were not appropriate. mainly used to assess the toxicity of pure metals (Van Straalen et al., 1989; Crommentuijn et al., 1995, 1997; Sandifer and 2.2. Collembola cultures Hopkin, 1996; Scott-Fordsmann et al., 1997; Smit and Van Gestel, 1998; Crouau et al., 1999; Fountain and Hopkin, 2001) or of A culture of F. candida was reared in the laboratory at 2071 1C in darkness, organic chemicals (Herbert et al., 2004; Eom et al., 2007). in glass containers with a base of plaster of Paris/charcoal powder mixture Application to complex mixtures such as polluted soils or wastes (ratio 4/1). Distilled water and a small amount of dried Baker’s yeast (as a food are less numerous (Smit and Van Gestel, 1996; Fountain and source) were added weekly. F. candida juveniles were collected two times a week Hopkin, 2004; Crouau et al., 2002; Crouau and Cazes, 2005; with a suction apparatus in order to select synchronized populations. Crouau and Pinelli, 2008). The effects of temperature (Snider and Butcher, 1973), pH and soil moisture (Holmstrup, 1997; Van 2.3. Toxicity tests Gestel and Van Diepen, 1997; Van Gestel and Mol, 2003) as well as the variability between strains on test sensitivity have been The test consists in exposing juveniles to field soils contaminated by Cd and in comparing reproduction, growth and mortality with those of animals placed in studied (Crommentuijn et al., 1995; Chenon et al., 1999; Crouau non-contaminated control soil. For toxicity tests, plastic containers of approxi- and Moia, 2006). mately 100 mL were used. The natural soils were spiked with Cd(NO3)2 (0, 50, 100, À Organic matter and pH have often been identified as key 200, 400 mg Cd g 1 dry soil) and were equilibrated for a week before starting the factors, which control the complexation and availability of metals, toxicity tests. Cd was dissolved in distilled water, which was used for soil moistening. For each soil and each spiking concentration, 8 test containers and thus influence metal toxicity, noticeably for Cd (see as (12 for the control) were filled with 45 g of moist soil. Fifteen F. candida juveniles example, Crommentuijn et al., 1997; Son et al., 2007). (8–12 days old juveniles) were introduced into each container. Indeed, we used To ensure their protection regarding agriculture and atmo- 8–12 days old animals rather than 10–12 days (as recommended by the ISO spheric inputs, CCs of HM must be determined for different kinds guideline no. 11267). Young animals are more sensitive to metal toxicity and thus of soils with emphasis on Collembola as target living organisms. we think that animals as young as possible must be tested. The containers were aerated twice a week. Exogenous yeast was not added during testing in order to be With that aim, processes and key factors of metal toxicity in closer to field conditions. The duration of exposure was 50 days, instead of the 28 ‘‘natural’’ soils must be better investigated. French soils present a days recommended in the ISO guideline (ISO no. 11267, 1998) in order to large variety of conditions and show pollution by HM, noticeably counterbalance the lack of added food during the assay. Indeed, this might lead to Cd and Pb (Hernandez et al., 2003; Probst et al., 2003a, b). a lower reproduction rate than those of normalised assays with food addition. Moreover, a longer exposure duration increases the sensitivity of the assay In this paper, we used the reproduction of F. candida as an (EC50-repro for Cd after 6 weeksoEC50-repro after 4 weeks; Van Gestel and Mol, indicator of soil ecosystem sensitivity to cadmium and to 2003), decreases the variability of test results, and therefore increases the determine CCs in soil solution. Moreover, we studied the efficiency of the assay (Crouau and Cazes, 2003). Lastly, a longer exposure duration mortality and the growth of F. candida to improve the under- was more realistic because in field conditions, Collembola are exposed to standing of Cd effects. To be as representative as possible of pollutants for longer than 28 days. Moreover, increasing the assay duration allows to increase the number of individuals at the end of the assay and, on natural conditions, three spiked typical natural French soils with account of this, counterbalances the effect of none food addition, which would different pH and OM characteristics were used to perform toxicity have decreased the number of individuals. 7 1 tests. In these particular conditions, Cd LC50 (lethal concentra- At the end of the exposure time (20 1 C, in darkness), containers were flooded with deionised water and gently stirred in order to make all living animals tions), LOEC (lowest efficient concentrations), EC50 and EC5 to float at the water surface. The water surface of each container was (efficient concentration) for reproduction were computed. Cd photographed. Adults and juveniles were counted and their lengths (from the concentrations in organisms and in soil solution and CaCl2 end of the posterior abdominal segment to the anterior margin of the head) were exchangeable Cd, were determined. The influence of the main measured. All the animals were measured, but the animal number depended on soil parameters (pH, OM) on toxicity, was assessed. Finally, a the series and was the lowest for highest Cd concentration series. The limit length method for estimating critical concentration from EC values is between juvenile and adult populations was defined as the length class presenting 5 the lowest individual abundance (sum of the five concentration abundances), presented. around the third quartile value. Wilcoxon’s two-sample test was used to assess significant differences of reproduction, mortality and length between exposed and control series. Non- 2. Materials and methods parametric methods were used given the non-normal distribution of the data and

the heterogeneity of variances. The EC5 and EC50 values were calculated by using the maximum likelihood-probit procedure (Toxcalc 5.0 software, Ives, 1996; EPA 2.1. Soil characteristics methods).

Experiments were performed using three soils : two cultivated soils from the South-West area of France (Aurade´, AU and Saint-Victor, SV) and a forest soil (EPC, 2.4. Chemical analyses under spruce cover) from the centre part of France were chosen (Table 1). These soils were selected given their varying pH and OM contents (Table 1), to For each spiked concentration, Collembola were extracted (alive) and placed investigate the inﬂuence of these two parameters on F. candida reproduction. The during 3 days in fasting containers with only a humid ﬁlter paper to control two cultivated soils have a very low OM content and are circumneutral to basic hygrometry. Collembola were then killed by freezing (À80 1C). For each soil and

Table 1 Physico-chemical properties of soils used for toxicity tests on reproduction of F. candida (OM: organic matter content; RMQS: French national network for soil quality measurement; RENECOFOR: French national network for forest ecosystems long-term survey, Ponette et al., 1997, data from Hernandez et al., 2003).

À1 Soils Origin Soil cover Clay (%) pH (H2O) OM (%) Cd (lg g )

AU Experimental catchment (SW France) Wheat/sunﬂower 37.2 8.2 2.0 0.29 EPC RENECOFOR (W France) Forest 19.4 4.5 16.5 0.10 SV RMQS (SW France) Corn 24.8 6.1 1.6 0.17 ARTICLE IN PRESS

each Cd test concentration, pools of 1779 mg dry weight (about 20 animals) were 3. Results digested (ultrapure HNO3, 90 1C) together with blank and standard samples in a clean room. Cd concentration measurements in Collembola were controlled by using the international reference material TORT-2 (Lobster Hepatopancreas). 3.1. Collembola and soil analyses After dissolution procedure, a single analytical measurement was done in a series including blank and reference material. Cd concentrations were analysed in 3.1.1. Cd concentrations in F. candida Collembola with a Perkin-Elmer inductively coupled plasma–mass spectrometer Cd concentrations in body have been measured for all the (ICP–MS). In soil solutions and in exchangeable CaCl solutions, Cd was analysed 2 À1 by inductively coupled plasma-optical emission spectrometry (ICP-OES). The main experiment concentrations, except for 400 mg g condition Cd isotope (114Cd) was measured as it presented the lowest deviation during ICP– because not enough animals were available for analytical À À MS measurements. The detection limit of Cd measurements was 1.6 Â 10 3 mg L 1 determination (Fig. 1). Cd concentrations were high in all À1 for body concentration and 1 mg L for water concentration. The significance of exposed animals. The highest Cd concentrations were found in analytical interference was negligible (e.g. As and Co levels were lower with À1 several orders of magnitude than Cd in spiked bodies and soil solutions). Collembola of the SV soil (except for the 200 mg g condition) À1 À1 ICP–MS and ICP-OES measurements were calibrated using a set of gradually with a maximum value of 1100 mg g dry wt for the 100 mg g concentrated external standards. For body concentration measurements, dry soil condition. The lowest concentration was measured in the 3 calibration lines were performed for a 14 sample set analysis, using 4 standards Collembola of the AU soil, except for the 200 mg gÀ1 condition for for each (with Cd content from 0 to 55 mg LÀ1). The correlation coefficient of the which the organisms concentrations in EPC and SV decreased. calibration line for 114Cd was very elevated (r=0.9999), which guaranties the quality of Cd measurement in bodies. During ICP–MS measurements, as a routine procedure, an In/Re mix was used as an internal standard for body concentration 3.1.2. Soil solution and exchangeable Cd concentrations, pH and DOC measurements. Dissolved organic carbon (DOC) in soil solutions was analysed using a Cd concentrations in the soil solutions and in the exchangeable Shimatzu TOC 5000 Carbon Analyser. The detection limit of DOC measurements fractions are summarised in Table 2. At the beginning of the assay was 0.1 mg LÀ1. Pore water was extracted by soil centrifugation (2000g, 15 min). (t1), soil solution concentrations increased with Cd nominal soil À2 À1 Exchangeable Cd was obtained by adding 10 mol L CaCl2 (soil/solution=1/10, concentrations for the three soils. But, concentrations in the soil w/w). Solutions were filtered (with 0.22 mm filter) before element and DOC solutions and in the exchangeable fractions were much higher in analyses. the EPC and SV soils than in the AU soil. Cd in soil solution increased with time for SV and decreased for EPC. 10000 Soil solution pH was circumneutral for AU and acidic for EPC and SV; pH was similar at t1 and t8 for AU, but it decreased AU EPC SV À 1000 slightly (half a unit) for EPC (except at 400 mg g 1) and very much for SV. EPC showed DOC concentrations higher than those of AU µ g.g-1) 100 and SV. This difference can be explained by the higher OM content of EPC (16.5% for EPC, 2% and 1.6%, respectively, for AU and SV). 10 DOC did not change very much during the assay in AU, whereas it increased in EPC and decreased in SV for the highest Cd 1 concentrations. Cd Collembola (

0.1 3.2. Collembola length 0 50 100 200 -1 Cd contamination in soils (µg.g ) AU soil: Two groups of distinct lengths were observed for each soil concentration. The individuals added to the test containers at Fig. 1. Cd concentrations in animals for the experimental concentrations (50, 100 and 200 mg gÀ1) applied on the three considered soils (AU, EPC and SV), compared the beginning of the assay are the longest animals. The shortest to the control Cd content in soils. (y axis: log unit). group was composed of juveniles, which were born during the

Table 2

Cd concentrations in the three soils (nominal value, dry wt. basis), in corresponding soil solution (at t0 +1 week and t0 +8 weeks) and in CaCl2 exchangeable fraction. pH(CaCl2) and pH(H2O) and DOC (at t0+1 week and t0+8 weeks) are also indicated.

À1 À1 À1 Cd soil Cd solution (mg mL ) Cd CaCl2(mg mL ) pH soil solution pH DOC (mg mL )

À1 Nominal (mg g ) t1 t8 t1 t8 CaCl2 t1 t8

AU 0 oDL oDL 0.05 7.11 7.72 6.9 26 23 50 0.003 oDL 0.24 7.25 7.65 7.04 25.6 20.2 100 0.009 oDL 0.58 7.56 7.57 6.98 27.1 20.3 200 0.47 0.43 2.07 7.45 7.45 7.03 31.4 30.1 400 1.69 2.93 9.95 7.09 7.4 6.94 31.7 33

EPC 0 0.02 0.01 0.11 4.09 3.73 2.98 14.3 21.2 50 10.60 5.21 29.85 3.85 3.58 2.96 24.5 24.9 100 30.02 19.80 67.11 3.89 3.73 3.01 25.8 43.7 200 90.54 45.28 112.65 3.4 3.1 3.05 26.3 66.2 400 280.29 195.06 284.59 3.4 3.49 2.98 40 94.2

SV 0 oDL 0.04 0.67 4.16 2.57 4.44 30.2 28.8 50 6.68 12.91 27.28 4.2 3.53 4.44 32.3 49.3 100 18.45 35.04 47.33 4.05 3.3 4.47 43.2 26.1 200 51.89 82.33 112.97 3.99 3.36 4.57 66.4 48.6 400 143.13 185.23 264.47 3.91 3.19 4.64 72.7 59 oDL: inferior to detection limit. ARTICLE IN PRESS

18 14 16 Cd0 12 14 Cd50 10 12 Cd100 10 Cd200 8 8 Cd400 6 6 4 4 Mean animal number

Mean animal number 2 2 0 0 1 1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.1 1.2 1.3 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.1 1.2 1.3 Length (mm) Length (mm)

5 4.5 AU EPC SV 4 0.8 3.5 3 0.6 2.5 *** 2 * 0.4 *** ** 1.5 1 Juvenile length(mm) Mean animal number 0.5 0.2 0 1 0 0 0 50 50 50 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.1 1.2 1.3 100 200 400 100 200 400 100 200 400 Length (mm)

Fig. 2. Number of juveniles and adults for each length class and each Cd soil concentration in AU (a), EPC (b) and SV (c) soils. (d) Medians, quartiles and mode of the lengths of juveniles (signiﬁcant differences to control: npo0.05; nnpo0.01; nnnpo0.001)

Table 3 signiﬁcant difference between the lengths of the blank series Mean and range (in brackets) EC5, EC50 and LOEC for reproduction (juveniles and of the other series (adults and juveniles groups) except for the number) and mortality (adults number) for the three soils (AU, EPC, SV) relatively À1 À 100 mg g series (Fig. 2d). to Cd concentrations in soil (nominal concentration, mg g 1), in soil solution and to À1 EPC soil: The maximum number of individuals was observed CaCl2 exchangeable Cd(mg mL ). for the 0.65 mm length. Contrary to what was observed for the AU

EC5 EC50 LOEC and SV soils, Collembola length in EPC soil showed a single population distribution for all tested Cd concentrations (Fig. 2b). Juveniles Applying the same method as for the other two soils, the limit AU Soil 43.5 (11–77) 182 (134–254) 400 (po0.05) between juveniles and adults was 0.85 mm. The increase of the Soil solution 7 Â 10À4 (5 Â 10À6–5 Â 10À3) 0.17 (0.05–0.6) 1.7 (po0.05) soil Cd concentration did not change significantly the median of Exchangea. 0.13 (0.012–0.32) 1.9 (1.1–3.8) 9.9 (po0.05) juvenile length, except for the 200 mg gÀ1 series. EPC SV soil: Collembola of the SV soil showed two populations with Soil 56 (26–72) 111 (96–133) 100 (po0.05) Soil solution 9 (2.6–13.4) 22 (18–28) 30 (po0.05) different length ranges (Fig. 2c) for all series. The limit between Exchangea. 42 (19–52) 72 (63–83) 67 (po0.05) juveniles and adults was 0.7 mm. Significant length differences of À SV the blank series were only detected for the 100 mg g 1 dry soil Soil 15 107 200 (po0.01) series (Fig. 2d). The lack of significant differences for the highest Soil solution 2.6 36 52 (po0.01) À Cd concentrations (200 and 400 mg g 1) could be due to the Exchangea. 8 57 113 (po0.01) higher variability and the less numerous data for these two series. Adults AU Soil ––– 3.3. Collembola reproduction and mortality Soil solution ––– Exchangea. ––– EPC The EC5, EC50 and LOEC for reproduction and mortality o Soil 110 (47–137) 16 (118–177) 200 (p 0.05) tests are given in Table 3. Comparison of the EC50-repro of the Soil solution 22 (8–29) 35 (24–39) 45 (po0.05) three soils on the basis of soil concentrations indicated that Cd Exchangea. 72 (38–85) 95 (76–103) 77 (po0.05) has the same toxicity for the SV and EPC soils and a lower toxicity SV Soil 78 (29–113) 223 (178–280) 200 (po0.05) for the AU soil. It was the opposite for the EC50-repro calculated on Soil solution 26 (8–42) 93 (71–121) 82 (po0.05) the basis of soil solutions. A decrease in juveniles (effect on Exchangea. 35 (12–55) 129 (98–170) 113(po0.05) reproduction—Fig. 3a) and in adults (effect on mortality—Fig. 3b) populations with increasing Cd concentrations was observed for the EPC and SV soils. The AU soil showed a decrease in the assay (Fig. 2a). The mean lengths of the two groups (juveniles and juvenile number, but not in the adult number. A small increase in adults) were 0.45 and 1 mm, respectively. The limit point adults and juveniles was often observed for the lowest Cd between the two populations was 0.75 mm. There is no concentrations, except in SV; the EPC and AU soils showed an ARTICLE IN PRESS

a b 14 90 AU AU 80 12 EPC EPC SV 70 SV 10 60 8 50

40 6 30 adults number 4 Juveniles number 20 2 10 0 0 0 50 100 200 400 0 50 100 200 400 Cd (µg g-1) Cd (µg g-1) Fig. 3. Dose–effect relationship of Cd (nominal concentrations in soil) on the reproduction ((a) mean number of juveniles7S.E.M.) and the mortality ((b) mean number of adults7S.E.M.) of F. candida.

À1 increase in juvenile number for the 50 mg g series, followed by a on reproduction was partly due to mortality (LOECmortality=200 regular decrease for the three highest concentrations. mg gÀ1;r= À0.57; po0.01). Cd might delay juvenile growth for 200 and 400 mg gÀ1 series (Fig. 2c). In contrast to the AU soil, for which juveniles were much more abundant than adults, animal 4. Discussion numbers were quite similar between juvenile and adult groups for the SV soil. A low juvenile number was also found in control Juvenile productions were low when compared to litterature containers indicating that the SV soil did not favour F. candida references (Van Gestel and Mol, 2003; Greenslade and Vaughan, reproduction. The low reproduction observed for the SV soil was 2003) and the results of the assays did not fulfil validity criteria of not related to an indirect effect on mortality since more adults the norm, probably due to differences in experimental conditions were counted in control containers for SV among the three soils. (no food addition, natural vs. artificial soil). Indeed, the objectives The low reproduction rate can be explained by the low OM were not to strictly follow the norm but to fit as much as possible content of the SV soil, which inhibits microbial activity and food the field conditions. The lowest juvenile production in control production for F. candida. series was observed for the SV soil (average of 1473 juveniles), Contrary to the AU and SV soils, juvenile and adult groups which was nearly neutral. The two other soils are either cannot be distinguished on length histogram of the EPC soil. The calcareous (AU) or have a high organic matter content (EPC), right part of the juvenile peaks (corresponding to the longest which seems to increase the reproduction rate. The decrease in juveniles) was confounded with the left part of the adult peaks juveniles numbers in the SV and EPC soils (200 and 400 mg gÀ1 (shortest adults). Nevertheless, a separation between juvenile and series) could be partly due to the effect of Cd on mortality of adult populations was defined at 0.85 mm according to the adults (Fig. 3). procedure previously used for AU and SV. The overlap of juvenile

Values of EC50 based on reproduction (EC50-repro) and adult populations can be attributed to the high OM content in were 182, 111 and 107 mg gÀ1 dry soil for AU, EPC and SV, this soil: it enhanced growth rate of initially introduced respectively. These values were consistent with some individuals and caused consequently an early laying and a quick studies (Crommentuijn et al., 1997;Crouau et al., 1999; Herbert growth of produced juveniles. et al., 2004), but two studies (Van Gestel and Van Diepen, Cd concentrations in organisms were lower for the AU soil 1997; Van Gestel and Mol, 2003) reported values which are than for the other soils at nominal Cd soil concentrations of 50 À1 about two times lower than the range of this study. High pH value and 100 mg g . AU also had the lowest EC50-repro on the basis of (AU) increased EC50 values based on nominal concentrations. nominal concentrations. These results were consistent with the À1 The EC5-repro were 43, 56 and 15 mg g dry soil for AU, EPC lowest soluble and exchangeable Cd concentrations found for the and SV, respectively. The 95% confidence interval for AU over- AU soil, if we consider that Cd measured in solution is lapped that of EPC, indicating that the two values are almost representative of bioavailable Cd (Van Gestel and Koolhaas, equivalent. 2004). However, several results did not corroborate this hypoth- For the AU soil, juvenile numbers decreased strongly with the esis: (1) body concentrations for EPC and SV with 200 mg gÀ1 increase in Cd concentration (Fig. 3b). Mortality of adults was nominal Cd were lower than for 50 and 100 mg gÀ1 exposures, low whatever the soil Cd concentrations. We did not observe whereas Cd in solution was 2 or 3 times higher; (2) it is a significant shift towards smaller lengths of juveniles with noteworthy that EC50-repro values based on total soil concentra- Cd concentration increase; therefore, in AU soil, Cd affected tions were very close among the three soils whereas exchangeable reproduction rather than mortality and growth of F. candida. Such Cd and Cd in solution were very different (more than 3500 times a higher sensitivity of the reproduction parameter by comparison higher for EPC50 gm gÀ1 and EPC100 gm gÀ1 soil solution at t1 than for to mortality and growth parameters was previously observed AU50 gm gÀ1 and AU100 gm gÀ1 and more than 110 times higher for (Van Straalen et al., 1989; Scott-Fordsmann et al., 1997; Fountain EPC50 gm gÀ1 and EPC100 gm gÀ1 exchangeable Cd than for the equiva- and Hopkin, 2001). lent with AU soil). These last differences can probably For the SV soil, only the number of juveniles in the 200 mg gÀ1 be attributed to the effect of pH since it is well known that series differed significantly from the blank (po0.01). This effect Cd availability in solution decreases with increasing pH (Van ARTICLE IN PRESS

Gestel and Mol, 2003). The results showed that Cd concentrations soil pH is very elevated (as in AU soil), but cannot occur in the EPC in soil solution and exchangeable fraction were not the only and SV soils with a lower pH. determining parameters for Cd toxicity on F. candida. Sources Fig. 4 shows the ratio of Cd concentrations in Collembola on Cd other than Cd contained in pore water must be considered to concentrations in pore water for the different nominal soil explain the results: (1) soil fungi are able to acidiﬁcate their close concentrations. Cd accumulation by F. candida relative to pore environment and to increase nutrient absorption, causing at the water concentration decreases with nominal concentration but is same time the solubilisation and absorption of trace elements always higher for AU soil.

(Bago et al., 1996; Casarin et al., 2003; Gadd, 2007; Finlay, 2008; EC50 and EC5 give informations on Cd effects on F. candida. Van Scholl et al., 2008). When these organisms are consumed by However, the ﬁnal object of that study is the evaluation of the Collembola, they represent a privileged route for trace metal contamination level compatible with the protection of almost all absorption. This phenomenon should be more important for a the soil invertebrates community. We have kept the reproduction carbonated soil as AU and would balance the very low Cd parameter because it is the most sensitive and ecologically concentrations in pore water. (2) Microorganisms consumed by relevant one. (Crommentuijn et al., 1995). Ideally, it would be F. candida constitute a ﬁrst step in metal internalisation which necessary to test Cd effect on almost all the soil arthropods leads to Cd ingested concentrations that are different from pore species. It is clearly impossible and we must try to correct the ECx water Cd concentrations. As an example, Posthuma (1992) found that we get for F. candida on the basis of the relative sensitivities that Cd concentrations in soil algae were equivalent to Cd of F. candida and of the other soil invertebrates for which tests concentrations in bulk soil for low values (0.009 mg gÀ1) but have been done. With this aim in mind, we tried to evaluate the were lower when bulk soil concentrations increased (0.046 mg Cd concentration preserving 95% of the reproduction of 95% of soil gÀ1 in algae for 0.56 mg gÀ1 in soil). The Cd accumulation rate of invertebrates (critical concentrations (CC)). Several authors con- these organisms would be more important for the AU soil than for sider that F. candida is less sensitive to trace metal than other the two other soils, relatively to pore water concentrations. (3) species of the pedofauna (Lubben, 1989; Greenslade and Vaughan,

Soil and exchangeable solutions were sieved with a 0.22 mm ﬁlter 2003; Son et al., 2007). EC50-repro for Cd of some soil invertebrates before analysis. This step eliminates Cd from solution when linked are given in Table 4. to suspended particles with diameter higher than 0.22 mm but The EC50-repro and LC50 of soil invertebrates were approximated which could be absorbed by Collembola (organic particles or with the log-normal distribution applied to the data of Table 4. It clays). When such particles are consumed, Cd would be partly allowed to quantify the sensitivity of F. candida in comparison desorbed from particles in Collembola digestive system due to with other soil invertebrates. This model has previously been used lowest pH (pH 6; Humbert, 1974) and consequently it would be to study species distribution of Collembola in natural commu- absorbed through intestinal epithelium. A decrease of one pH unit nities (Syrek et al., 2006) and NOEC distribution (Aldenberg and reduces Cd sorption by about 75% (Temminghoff et al., 1995). Cd Slob, 1993). We get 20 and 37, respectively, for the EC50-repro and adsorbed onto these particles was not measured as Cd in soil LC50 of the 5% more sensitive species. On this basis, F. candida solution. The process described above can only be observed when would be about 9 times less sensitive for reproduction and 33 times for lethality than the 5% most sensitive group. So, the difference is larger for lethality than for reproduction. This 10000 AU observation is in agreement with the high sublethal sensitivity EPC index (ratio between the lethal effect concentration and the solution 1000 SV sublethal effect concentration—SSI) previously found for F. candida (Crommentuijn et al., 1995 ). F. candida put a higher

/ [Cd] 100 priority on survival than on reproduction. Applying the correction factors dealing with reproduction, the CC of the AU, EPC and SV Â À5 À1 10 soils would be, respectively, 8 10 , 1.1 and 0.3 mg mL . These collembole values can be compared to critical concentrations obtained

[Cd] according to the method exposed in De Vries et al. (2007), which 1 0 50 100 200 400 is based on modelling, using pH and MO content as prediction parameters: 1 Â 10À4, 1.7 Â 10À3, 5 Â 10À4 mg mLÀ1, respectively, Fig. 4. Cd bioaccumulation factor relatively to pore water concentrations. for AU, EPC and SV. The values computed in this study are close to

Table 4

EC50-repro and LC50 calculated on the basis of Cd soil concentrations for 12 soil invertebrates. (Bold types characters correspond to mean of several data.)

EC50-repro LC50 Reference

(Cd soil) (Cd soil)

Aporrectodea caliginosa 35 540 Khalil et al. (1996) Caenorhabditis elegans 627 Donkin and Dusenbery (1994) Enchytraeus albidus 115 364 Lock and Janssen (2001) Eisenia andrei 33 417 Van Gestel et al. (1993) Eisenia fetida 116 1095 Spurgeon et al. (1994), Spurgeon and Hopkin (1995), Neuhauser et al. (1985), Fitzpatrick et al. (1996), Lock and Janssen (2001) Lumbricus terrestris 256 Fitzpatrick et al. (1996) Folsomia candida 180 1213 Van Gestel and Van Diepen (1997), Crommentuijn et al. (1995, 1997), Sandifer and Hopkin (1996, 1997), Lock and Janssen (2001) Plectus acuminatus 321 Kammenga et al. (1996) Paronychiurus kimi 60 90 Son et al. (2007) Proisotoma minuta 125 Nursita et al. (2005) Sinella coeca 28 12 Menta et al. (2006) Sinella communis 50 374 Greenslade and Vaughan (2003) ARTICLE IN PRESS

the values determined by the mean of de Vries et al. method for Crouau, Y., Cazes, L., 2003. What causes variability in the Folsomia candida AU but greatly higher for EPC and SV. Indeed, it means that for reproduction test. Appl. Soil Ecol. 22, 175–180. Crouau, Y., Cazes, L., 2005. Unexpected reduction in reproduction of soils with a significant proton and organic matter content (EPC Collembola exposed to an arsenic-contaminated soil. Environ. Toxicol. Chem. and SV), the model from De Vries et al. based on these parameters 24 (7), 1716–1720. does not give suitable results. Thus, the apparent adequacy for Crouau, Y., Moia, C., 2006. The relative sensitivity of growth and reproduction in critical limits obtained for carbonate soil (AU) might be the springtail, Folsomia candida, exposed to xenobiotics in the laboratory: an indicator of soil toxicity. Ecotoxicol. Environ. Saf. 64, 115–121. coincidental. Crouau, Y., Pinelli, E., 2008. Comparative ecotoxicity of three polluted indus- Our results state that comparing critical limits determined by trial soils for the Collembola Folsomia candida. Ecotoxicol. Environ. Saf. 71, direct toxicological effect on soil organisms with those predicted 643–649. Dambrine E., Probst A., Party J.P., 1993. De´termination des ‘‘charges critiques’’ de by models based on soil parameter remain still difficult. Further polluants atmosphe´riques pour les ećosystemes naturels, en particulier investigations are needed. forestiers. Bases theóriques - Projet d’application au cas des Vosges. Pollution Atmosphe´rique, No. spećial Pollution de l’Air et Charges Critiques, juin, 1993, pp. 21–28. De Vries, W., 1988. Critical deposition levels for nitrogen and sulphur on Dutch 5. Conclusion forest ecosystems. Water Air Soil Pollut. 42, 221–239. De Vries, W., Reinds, G.J., Posch, M., 1994. Assessment of critical loads and their This study confirms that Collembola reproduction is a more exceedances on European forests using a one-layer steady-state model. Water Air Soil Pollut. 72, 357–394. sensitive parameter than mortality and growth regarding Cd De Vries W., Bakker D.J., 1996. Manual for calculating critical loads of heavy metals toxicity. It also shows that low organic matter content is a limiting for soils and surface waters. S.C-DLO Publishers, report 114, 173 pp. factor for reproduction. Sensitivity to Cd was enhanced for low OM De Vries, W., Lofts, S., Tipping, E., Meili, M., Groenenberg, J.E., Schutze,¨ G., 2007. content and acidic pH. We succeed in determining EC for Cd with Impact of soil properties on critical concentrations of cadmium, lead, copper, 50 zinc and mercury in soil and soil solution in view of ecotoxicological effects. three different natural soils and we show that effect of Cd on Rev. Environ. Contam. Toxicol. 191, 47–89. reproduction is better described by soil or body concentrations than Donkin, S.G., Dusenbery, D.B., 1994. Using the Coenorhabditis elegans soil toxicity by soil solution concentration. In spite of different Cd concentra- test to identify factors affecting toxicity of four metal ions in intact soil. Water Air Soil Pollut. 78, 359–373. tions in soil solutions, EC50 are rather similar. The critical limits Eom, I.C., Rast, C., Veber, A.M., Vasseur, P., 2007. Ecotoxicity of a polycyclic determined in this study do not really fit those given by soil aromatic hydrocarbon (PAH)-contaminated soil. Ecotoxicol. Environ. Saf. 67 parameter modelling, which implies further investigations on (2), 190–205. coupled model particularly considering the influence of particulate Finlay, R., 2008. Ecological aspects of mycorrhizal symbiosis: with special emphasis on the functional diversity of interactions involving the extraradical matter and microorganisms. mycelium. J. Exp. Bot. 59, 1115–1126. Fitzpatrick, L.C., Muratti-Ortiz, J.F., Venables, B.J., Goven, A.J., 1996. Comparative toxicity in earthworms Eisenia fetida and Lumbricus terrestris exposed to Acknowledgments cadmium nitrate using artificial soil and filter paper protocols. Bull. Environ. Contam. Toxicol. 57, 63–68. Fountain, M.T., Hopkin, S.P., 2001. Continuous monitoring of Folsomia candida This study benefited from the financial aids from ADEME, (Insecta: Collembola) in a metal exposure test. Ecotoxicol. Environ. Saf. 48, CNRS, INRA and Toulouse University. 275–286. Fountain, M.T., Hopkin, S.P., 2004. Biodiversity of Collembola in urban soils The authors thank to A. Canut, A. Magro, F. Candaudap and and the use of Folsomia candida to assess soil ‘‘quality’’. Ecotoxicology 13 (6), A. Alric, for their help in field sampling, sample preparation, or 555–572. analysis and to K. Dexter for checking the English. Gadd, G., 2007. Geomycology: biogeochemical transformations of rocks, minerals, metals and radionuclides by fungi, bioweathering and bioremediation. Mycol. T. Bur benefited from a joint fellowship Agence de l’Envir- Res. 111, 3–49. onnement et de la Maˆıtrise de l’Energie (ADEME)–Institut Greenslade, P., Vaughan, G.T., 2003. A comparison of Collembola species for National de la Recherche Agronomique (INRA)/Re´seau de Mesure toxicity testing of Australian soils. Pedobiologia 47, 171–179. de la Qualite´ des Sols (RMQS). L. Gandois benefited from a ADEME Herbert, I., Svendsen, C, Hankard, P, Spurgeon, D., 2004. Comparison of instantaneous rate of population increase and critical-effect estimates fellowship. in Folsomia candida exposed to four toxicants. Ecotoxicol. Environ. Saf. 57, 183–193. Hernandez, L., Probst, A., Probst, J.L., Ulrich, E., 2003. Heavy metal distribution in References some French forest soils: evidence for atmospheric contamination. Sci. Total Environ. 312 (1–3), 195–219. Hettelingh, J.P., De Vries, W., Schopp,¨ W., Downing, R.J., De Smet, P.A.M., 1991. Aldenberg, T., Slob, W., 1993. Confidence limits for hazardous concentrations based on logistically distributed NOEC toxicity data. Ecotoxicol. Environ. Saf. Chap. 3. Methods and data. in : Mapping Critical loads for Europe. In: 25, 48–63. Hettelingh, J.P., Downing, R.J., De Smet, P.A.M. (Eds.), CEC Technical Report 1. Bago, B., Vierheilig, H., Piche, Y., Azcon-Aguilar, C., 1996. Nitrate depletion and pH National Institute of Public Health and Environmental Protection, Bilthoven, changes induced by the extraradical mycelium of the arbuscular mycorrhizal pp. 1–86. fungus Glomus intraradices grown in monoxenic culture. New Phytol. 133, Holmstrup, M., 1997. Drought tolerance in Folsomia candida Willem (Collembola) 273–280. after exposure to sublethal concentrations of three soil-polluting chemicals. Baize, D., 1997. Teneurs totales en e´le´ments traces me´talliques dans les sols. INRA Pedobiologia 41, 361–368. Editions. Hornung, M, Bull, KR, Cresser, M, Hall, J, Langan, SJ, Loveland, P, Smith, C., 1995. An Casarin, V., Plassard, C., Souche, G., Arvieu, J., 2003. Quantification of oxalate ions empirical map of critical loads of acidity for soils in Great Britain. Environ. and protons released by ectomycorrhizal fungi in rhizosphere soil. Agronomie Pollut. 90, 301–310. 23, 461–469. Humbert, W., 1974. Etude du pH intestinal d’un Collembole (Insecta, Apte´rygote). Chenon, P., Gisclard, C., Crouau, Y., 1999. The use of Folsomia candida (Collembola, Rev. Ecol. Biol. Sol. 11, 89–97. Isotomidae) for the bioassay of xenobiotic substances and soil pollutants. J. ISO 11268-1, 1994. Soil quality—effects of pollutants on earthworms (Eisenia Appl. Soil Ecol. 12, 103–111. fetida). Part 1. Method for the determination of acute toxicity using artificial Crommentuijn, T., Doodeman, C, Vanderpol, J, Doornekamp, A, Rademaker, M., Van soil substrate. International Standard Organization, Geneve. Gestel, C., 1995. Sublethall sensitivity index as an ecotoxicity parameter ISO 11268-2, 1997. Soil quality—effects of pollutants on earthworms (Eisenia measuring energy allocation under toxicant stress—application to cadmium in fetida). Part 2. determination of effects on reproduction. International Standard soil arthropods. Ecotoxicol. Environ. Saf. 31, 192–200. Organization, Geneve. Crommentuijn, T., Doornekamp, A, Van Gestel, C.A.M., 1997. Bioavailability and ISO 11267, 1998. Soil quality—inhibition of reproduction of Collembola (Folsomia ecological effects of cadmium on Folsomia candida (Willem) in an artificial soil candida) by soil pollutants. International Standard Organization, Geneve. substrate as influenced by pH and organic matter. Appl. Soil Ecol. 5, 261–271. ISO 11268-3, 1999. Soil quality—effects of pollutants on earthworms (Eisenia Crouau, Y., Chenon, P, Gisclard, C., 1999. The use of Folsomia candida (Collembola, fetida). Guidance on the determination of effects in field situations. Interna- Isotomidae) for the bioassay of xenobiotic substances and soil pollutants. Appl. tional Standard Organization, Geneve. Soil Ecol. 12, 103–111. ISO 16387, 2001. Soil quality—effects of pollutants on Enchytraeidae (Enchytraeus Crouau, Y., Gisclard, C., Perotti, P., 2002. The use of Folsomia candida (Collembola, sp.)—determination of effects on reproduction and survival. Interational Isotomidae) in bioassays of waste. Appl. Soil Ecol. 19, 65–70. Standard Organization, Geneve. ARTICLE IN PRESS

Ives, M.A., 1996. In: TOXCALC 5.0. Tidepool Scientific Software, McKinleyville, CA, Sandifer, R., Hopkin, S., 1997. Effects of temperature on the relative toxicities of Cd, USA 80 pp. Cu, Pb and Zn to Folsomia candida (Collembola). Ecotoxicol. Environ. Saf. 37, Kammenga, J.E., Van Koert, P.H.G., Riksen, J.A.G., Korthals, G.W., Bakker, J., 1996. A 125–130. toxicity test in artificial soil based on the life-history strategy of the nematode Scott-Fordsmann, J.J., Krogh, P.H., Weeks, J.M., 1997. Sublethal toxicity of copper to Plectus acuminatus. Environ. Toxicol. Chem. 15, 722–727. a soil-dwelling springtail Folsomia fimetaria (Collembola: Isotomidae). Environ. Khalil, M.A., Abdel-Lateif, H.M., Bayoumi, B.M., Van Straalen, N.M., Van Gestel, C.A.M., Toxicol. Chem. 16, 2538–2542. 1996. Effects of metals and metal mixtures on survival and cocoon production of Slootweg, J., Hettelingh, J., Posch, M., Dutchak, S., Ilyn, I., 2005. Critical loads of the earthworm Aporectodea caliginosa. Pedobiologia 40, 548–556. cadmium, lead and Mercury in Europe. Working Group on effects of the Lock, K., Janssen, C.R., 2001. Cadmium toxicity for terrestrial invertebrates: taking soil convention on Long-range Transboundary Air Pollution. parameters affecting bioavailability into account. Ecotoxicology 10, 315–322. Smit, C.E., Van Gestel, C.A.M., 1996. Comparison of the toxicity of zinc for the Lofts, S., Spurgeon, D.J., Svendsen, C., Tipping, E., 2004. Deriving soil critical limits springtail Folsomia candida in artificially contaminated and polluted field soils. for Cu, Zn, Cd and Pb: a method based on free ion concentrations. Environ. Sci. Appl. Soil Ecol. 3, 127–136. Technol. 38 (13), 3623–3631. Smit, C.E., Van Gestel, C.A.M., 1998. Effects of soil type, prepercolation, and ageing Lubben, B., 1989. Influence of sewage sludge and heavy metals on the abundance on bioaccumulation and toxicity of zinc for the springtail Folsomia candida. of Collembola on two agricultural soils. In: Third International Seminar on Environ. Toxicol. Chem. 17, 1132–1141. Apterygota. R. Dallai, pp. 419–428. Snider, R.M., Butcher, J.W., 1973. The life history of Folsomia candida (Willem) Massabuau, J.C., Probst, A., Guerold, F., 1995. Critical loads of acidity to (Collembola, Isotomidae) relative to temperature. Great Lake Entomol. 6, streamwaters in the Vosges mountains (France): biological criteria. In: 97-106. Landmann, G., Bonneau, M. (Eds.), Forest Decline and Atmospheric Deposition Effects in the French Mountains. Springer, Berlin, Heildelberg, New York, pp. Son, J., Mo, H, Kim, J, Ryoo, M, Cho, K., 2007. Effect of soil organic matter content 387–393 461p.. and pH on toxicity of cadmium to Paronychiurus kimi (Lee) (Collembola). Menta, C., Maggiani, A., Vattuone, Z., 2006. Effects of Cd and Pb on the survival and J. Asia-Pac. Entomol. 10 (1), 55–61. juvenile production of Sinella coeca and Folsomia candida. Eur. J. Soil Biol. 42, Son, J., Ryoo, M., Jung, J, Cho, K., 2007. Effects of cadmium, mercury and lead on the 181–189. survival and instantaneous rate of increase of Paronychiurus kimi (Lee) Moncoulon, D., Probst, A., Party, J.-P., 2004. Weathering, atmospheric deposition (Collembola). Appl. Soil Ecol. 35, 404–411. and vegetation uptake: role for ecosystem sensitivity to acid deposition and Spurgeon, D.J., Hopkin, S.P., Jones, D.T., 1994. Effects of cadmium, copper, lead and critical load. C. R. Geosci. 336, 1417–1426. zinc on growth, reproduction and survival of the earthworm Eisenia fetida Moncoulon, D., Probst, A., Martinson, L., 2007. Modeling acidification recovery on (Savigny): assessing the environmental impact of point-source metal con- threatened ecosystems: application to the evaluation of the Gothenburg tamination in terrestrial ecosystems. Environ. Pollut. 84, 123–130. protocol. Water Air Soil Pollut. 7 (1–3), 307–316. Spurgeon, D.J., Hopkin, S.P., 1995. Extrapolation of the laboratory-based OECD Nilsson, J., Grennfelt, P., 1988. Critical loads for sulphur and nitrogen. Miljoreport earthworm toxicity test to metal contaminated field sites. Ecotoxicology 4, 15, 418. 190–205. Neuhauser, E.F., Loehr, R.C., Milligan, D.L., Malecki, D.L., 1985. Toxicity of metals to Syrek, D., Weiner, W.M., Wojtylak, M., Olszowska, G., Kwapis, Z., 2006. Species het earthworm Eisenia fetida. Biol. Fertil. Soils 1, 149–152. abundance distribution of collembolan communities in forest soils polluted Nursita, A., Ayulungit, I, Singh, A, Balwant, A., Lees, E., 2005. The effects of with heavy metals. Appl. Soil Ecol. 31, 239–250. cadmium, copper, lead, and zinc on the growth and reproduction of Proisotoma Temminghoff, E.J.M., van der Zee, S.E.A.T.M., De Haan, 1995. Speciation and minuta Tullberg (Collembola). Ecotoxicol. Environ. Saf. 60, 306–314. calcium competition effects on cadmium sorption by sandy soil at various pH. Party, J.P., Probst, A., Dambrine, E., Thomas, A.L., 1995. Critical loads of acidity to Eur. J. Soil Sci. 46, 649–655. France: sensitivity areas in the north-eastern France. Water Air Soil Pollut. 85 Van Gestel, C.A.M., Dirven-Van Breemen, E.M., Baerselman, R., 1993. Accumulation (1–4), 2407–2412. and elimination of cadmium, chromium and zinc and effects on growth Party, J.P, Probst, A., Thomas, A.L., Dambrine, E., 2001. Charges critiques d’acidite´ and reproduction in Eisenia andrei (Oligichaeta, Annelida). Sci. Total Environ., en polluants atmosphe´riques en France: conse´quences vis a vis des sols et des 585–597. ´ peuplements forestiers. Pollution Atmospherique 172 (oct-dec), 519–527. Van Gestel, C.A.M., Van Diepen, A.M., 1997. The influence of soil moisture content Ponnette, Q., Ulrich, E., Brethes, A., Bonneau, M., Lanier, M., 1997. RENECOFOR - on the bioavailability and toxicity of cadmium for Folsomia candida Willem Chimie des sols dans les 102 Peuplements du re´seau. Campagne des mesures (Collembola: Isotomidae). Ecotoxicol. Environ. Saf. 36, 123–132. 1993/1995. Office National des Foretsˆ Ed., De´partement des Recherches Van Gestel, C.A.M., Mol, S., 2003. The influence of soil characteristics on cadmium Techniques, 2-84207-100-X. toxicity for Folsomia candida (Collembola: Isotomidae). Pedobiologia 47, 387– Probst, A., Hernandez, L., Probst, J.L., 2003a. Heavy metals partitioning inthree French 395. forest soils by sequential extraction procedure. J. Phys. IV 107, 1103–1106. Probst, A., Moncoulon, D., Godde´ris, Y., Hernandez, L., Party, J.-P., 2003b. Critical Van Gestel, C.A.M., Koolhaas, J.E., 2004. Water-extractability, free ion activity, and loads for lead in France: first results on forest soils. J. Phys. IV France 107, pH explain cadmium sorption and toxicity to Folsomia candida. (Collembola) in 1111–1114. seven soil-pH combinations. Environ. Toxicol. Chem. 23, 1822–1833. Posthuma, L., 1992. Genetic ecology of metal tolerance in Collembola. Thesis, Vrije Van Scholl, L., Kuyper, T., Smits, M., Landeweert, R., Hoffland, E., Van Breemen, N., Universiteit, Amsterdam. 2008. Rock-eating mycorrhizas: their role in plant nutrition and biogeochem- Riepert, F., 1995. First report of the second international ring test on a method for ical cycles. Plant Soil 303, 35–47. determining the effects of chemicals or soil contaminants on the reproduction Van Straalen, N.M., Schobben, J.H.M., De Goede, R.G.M., 1989. Population of collembola. Report of the Biologische bundesanstalt fur¨ Land- und consequences of cadmium toxicity in soil microarthropods. Ecotoxicol. Forstwirtschaft Institut fur¨ Chemikalienprufung.¨ Environ. Saf. 17, 190–204. Sandifer, R., Hopkin, S., 1996. Effects of pH on the toxicity of cadmium, copper, lead Warfvinge, P, Sverdrup, H., 1992. Calculating critical loads of acid deposition and zinc to Folsomia candida Willem, 1902 (Collembola) in a standard with profile—A steady-state soil chemistry model. Water Air Soil Pollut. 63, laboratory test system. Chemosphere 33, 2475–2486. 119–143.